Javi and Chita

 

You know how I've been screaming at the political punditry for a few years now over their theory of the Latin voting bloc becoming more Republican over time as the community's inherent conservatism asserts itself, because

1. It's not a bloc, and
2. that's hiding what's happening

—and hiding from it, of course—it's about race. As with Allan Sherman's upwardly mobile young Jews in the early 1960s who

Traded their used MG
for a new XKE
switched to the GOP
that's the way things go

it's about the social construction of whiteness.

But I didn't really have any data for it, or know how to find it or whether it could possibly exist, until I saw this message from Jonathan Chait: 

Astonishing finding from @databyler https://t.co/U1GjarsYJv pic.twitter.com/IvuLDxsi5h

— Jonathan Chait (@jonathanchait) July 19, 2023

referencing a note in The Washington Post by David Byler which was in turn reporting some recent research by a Berkeley graduate student, Alexander Agadjanian, and colleagues, discovering a correlation between people who switched their presidential votes from Obama in 2012 to Trump in 2016 and people who switched races around the same period—who changed their answer to the question of racial identity in big surveys like those of the American National Election Studies.

I suppose one of the things that astonished Chait was the discovery that it's not too uncommon for Americans to do that; to change their minds about what race they belong to, though I should add that it was probably unusually common between 2000 and 2020, when the national census changed its approach to the question pretty severely three times. I had a pretty clear idea of that, anyway, which I wrote about in June 2022; but I wasn't quite prepared for the pattern with "Hispanics":

Those who'd classified themselves as Hispanic weren't nearly as promiscuous in changing their minds as the more ambiguous "Multiracial" and "Other" groupings, but much more so than whites, Blacks, and Asians; Hispanics who did reclassify themselves overwhelmingly decided they were white, while multiracials and others were all over the map. It's only the Hispanics who appear to be doing something systematic. 

That was the 2021 paper by Agadjanian and Dean Lacy. In the 2022 paper, Agadjanian moved on to the question of voting: 

 Across several models and robust to various controls, switching from a non-Republican vote in 2012 to a 2016 Republican vote (i.e., non-Romney to Trump) significantly predicts nonwhite to white race change. Among nonwhites who did not vote Republican in 2012, switching to a Republican vote in 2016 increases the probability of adopting a white racial identity from a 0.03 baseline to 0.49, a 1,533 percent increase. Individuals originally identifying as Mixed and Hispanic drive this identity-voting link.

And it only works in this direction. Important numbers of people decide they are Black or Mixed, but there's no particular association with their politics. The only pattern, and it's a very robust one, is: if you were a 2012 Obama voter in some doubt as to whether you were white or not, voting for Trump in 2016 helped you decide you were, or was at least part of the decision to identify as white.

The WaPo article takes this as something new on top of the much-heralded swing to the Republicans of the Latin community, and good news for the Republicans worried about how their white voter basis is doomed to minority status:

this data will come as good news for many in the GOP.... Many once-loyal Hispanic and Asian American Democrats voted for Trump in 2020. Combined with the fact that some non-White Democrats simply left their racial groups altogether in recent years, these data points suggest demography is not destiny.

But I would say the two are a single phenomenon—turning altogether white is just its most radical form. And it's not as significant for future elections as might be thought. For one thing, it seemed in 2022 to be largely over already, except in Florida, and for another it's intrinsically limited: to those light-skinned enough to "pass", for one thing, like the Miami Cubans who were white before they were exiled (and have been overwhelmingly Republican ever since anyway, associating the party with anti-communism), or the patricians of the Mexican border who were white for 300 years before the border existed, and associated with other indicators, like religious evangelicalism—Catholics and practitioners of Afro-Carribbean religions will continue to be Democrats, just as they continue to be brown.

And they will continue to be brown, apparently. The number of those who became white is absolutely dwarfed by the number who went the opposite way, especially in the South:

In 2010, the majority of the Hispanic population in the South reported their race as White alone (62.9%). By 2020, only 23.2% reported as White alone. All regions saw a decrease in White alone reporting for the Hispanic population over the decade, with the sharpest decline (nearly 40%) in the South.

I don't think Republicans have found themselves a convenient way of manufacturing white people yet.